In this section we look at the different labour market developments by demographic characteristics of the population in working age, focusing on gender (men and women), age (young aged 15-24, prime age (25-54), and old age 55-64) and education (low skilled - ISCO 1, 2 -, medium skilled - ISCO 3, 4 - and high skilled - ISCO 5, 6-).
1.6.1. Employment
Even if the crisis has hit the European labour market severely, different demographic groups have fared differently during the year 2009, as Table 6 shows. While the employment of men shrank by 2.7%, the employment of women fell by a comparatively small 0.7%. This stark gender difference in employment performance is not limited to the EU, but is also present in the US, where the employment of men declined by 4.9% while that of women only by 2.5%. Table 6 - Structure of Employment
2009
(in 1000s) 2008-2009Change 2007-2008Change Average Change 2000-2009
Total employment (age 15-64) 213,883 -1.8% 1.1% 0.8%
of which men 116,747 -2.7% 0.7% 0.4%
of which women 97,136 -0.7% 1.7% 1.4%
Young employment (age 15-24) 20,872 -7.5% -0.3% -0.9%
of which men 11,214 -9.1% -0.7% -1.0%
of which women 9,622 -5.9% 0.3% -0.7%
Prime age employment (age 25-54) 165,480 -1.7% 0.9% 0.6%
of which men 89,679 -2.5% 0.4% 0.2%
of which women 75,800 -0.8% 1.5% 1.2%
Old employment (age 55-64) 27,532 2.5% 4.1% 4.1%
of which men 15,854 1.2% 3.8% 3.3%
of which women 11,678 4.3% 4.4% 5.2%
Low skilled employment 47,845 -5.8% -2.5% -1.9%
of which men 28,003 -6.6% -2.3% -1.8%
of which women 19,842 -4.7% -2.8% -2.0%
Medium skilled employment 105,353 -2.4% 1.3% 1.3%
of which men 58,483 -2.8% 1.1% 1.0%
of which women 46,870 -1.8% 1.6% 1.5%
High skilled employment 60,128 2.8% 4.1% 3.8%
of which men 29,933 1.7% 3.0% 2.8%
of which women 30,196 3.9% 5.3% 4.9%
Source: Commission services.
The gender difference in employment performance during the crisis is generally explained by different distributions of men and women within industries. Men are more likely than women to be employed in industries heavily hit by the crisis, like construction and manufacturing, which are usually more reactive to the cycle. This is documented in Graph 13, showing on the left scale the share of men in total employees by industry. Industries are shown in ascending order of the share of male employees. At the lower end (left in the picture) stands the health care sector, with only 21% male employees, and at the upper
end (right in the picture) construction, where more than 90% are men. The right scale shows the employment growth between 2008 and 2009. It can be seen that the higher the share of males among the employees in an industry the more negative has been the employment development. Whereas employment actually grew by 3% in the female dominated health care sector, it shrank by 7% in male dominated construction. The exception is electricity, gas and water supply where 79% of total employees are men and which recorded an employment growth of 3% between 2008 and 2009.
There are significant differences in employment developments by age group, as there are by gender. The employment of the young (aged 15-24) shrank heavily (-7.5%), while that of the prime age group (25-54) declined by 1.7% and that of older people (55-64) increased, despite the crisis, by a considerable 2.5%. In 2009, as in most years before during the decade, the employment performance was the better the older the group was. One reason why the crisis hit the young so hard is the high share of temporary employment among them. In 2008, among the 15 to 24 years old 40% had temporary contracts, whereas only 11% of the 25 to 64 years old were temporarily employed. On the other end of the age distribution, the
actual rise in employment of old might reflect the tightening in the early retirement conditions. It is also conceivable that the crisis’ negative impact on the wealth of private households induced older employees to postpone retirement. Looking at the age dimension country by country (Table 7), the age pattern is the same in almost all countries but the level differs among them. The young do terribly badly regarding employment in Spain, Ireland and the Baltic countries, which are all heavily hit by the crisis. In these countries, young employment lost more than 20% in 2009. On the other hand, for the old an employment growth of more than 6% has been recorded in five countries (Luxembourg, Hungary, Poland, Slovenia, and Slovakia). Graph 13 - Share of Males and Employment Growth by Industry
Construction
Mining and Quarrying
Electricity
, gas and water supply
Transport and communication
Manufacturing
Agriculture and fishing
Public admin
istration
Business activities and real est
ate
Wholesale and retail trade
Financial intermediation
Hotels and restaurants
Education Health -100% -80% -60% -40% -20% 0% 20% 40% 60% 80% 100% Share of males 2008 Employment growth 2008-2009 -8% -6% -4% -2% 0% 2% 4% 6% 8% Share of males Employment growth
The educational attainment is a further dimension of employment which leads to remarkable performance differences. Low- skilled employment shrank by 5.8% in 2009. Medium-skilled employment fell by 2.4%. Only high-skilled employment grew during the crisis, by 2.8%. This shows that the skill upgrading in employment continued in 2009, confirming the previous long run trend. Of particular interest is the gender dimension of the employment growth of the high skilled. In 2009, the employment of high-skilled men grew by 1.7% and that of women by 3.9%, leading for the first time to higher female high-skilled employment (30.2 million) than male high- skilled employment (29.9 million) .
Table 8 shows the employment developments in 2009 broken down by its main components. The number of employees fell by 1.8%. The number of self-employed fell only by 0.4%. However, this does not mean the self-employed
do better during the crisis. The number of self- employed was already falling in 2008, while the number of employed was still growing. Full- time employment also shrank by 2.4%, while part-time employment grew by 1.1%, which shows that some of the adjustment in the total hours worked has come from a shift from full- to part-time.
Temporary employment dropped sharply between 2008 and 2009. The number of temporary employees fell by 5.9%. Due to their low employment protection, temporary employees accounted for a disproportionately high share of the decrease in the number of employees. While in 2008 14% of employees were employed with temporary contracts, temporary employees account for about 45% of the reduction in the total number of employees. Spain, with its pronounced dual labour market, provides an extreme example. In 2008, about 29% of the employees had a temporary contract. Table 7 - Employment Growth by Country and Age
Total (age 15-65) Young Prime age Old
Belgium -0.6% -6.9% -0.5% 4.3% Bulgaria -3.1% -8.6% -3.0% -0.2% Czech Republic -1.5% -6.2% -1.3% -0.6% Denmark -3.0% -1.2% -3.7% -1.2% Germany -0.3% -3.2% -0.8% 4.8% Estonia -9.1% -23.4% -8.6% -0.3% Ireland -8.8% -27.3% -6.4% -2.3% Greece -1.1% -4.9% -1.1% 0.7% Spain -6.8% -23.9% -5.6% -1.6% France -0.8% -2.4% -1.4% 4.7% Italy -1.6% -10.8% -1.7% 5.1% Cyprus -0.5% -7.3% -0.6% 4.9% Latvia -11.7% -28.3% -9.2% -10.3% Lithuania -6.9% -20.8% -5.8% -3.0% Luxembourg 6.4% 17.1% 4.6% 16.9% Hungary -2.5% -10.6% -3.1% 7.4% Malta 0.6% -3.0% 1.6% -0.6% Netherlands -0.3% -1.2% -1.2% 5.5% Austria -0.4% -2.7% -0.2% 0.6% Poland 0.5% -5.6% 0.3% 7.9% Portugal -2.8% -12.1% -2.1% -0.8% Romania -0.9% -3.1% -1.0% 1.6% Slovenia -2.1% -11.1% -2.0% 8.0% Slovakia -2.8% -15.1% -2.4% 6.1% Finland -3.0% -11.1% -2.4% 0.2% Sweden -2.3% -7.4% -1.8% -1.0% United Kingdom -1.7% -7.6% -0.8% -0.5%
1.6.2. Employment rate
The employment rate deserves a discussion on its own since it features prominently in the Europe 2020 strategy, as it did in the Lisbon strategy. The development of the employment rate mainly mirrors the development of employment, but is also influenced by (slower) movements of the population (the denominator of the employment rate) in the relevant age brackets. The Europe 2020 goal is a European wide employment rate of 75% among the 20 to 64 years old. Graph 14 shows the development of this indicator. In 2009, this indicator reached 69.1%, down by 1.3 pp. compared to 2008. The male employment rate fell by 2.1 pps. to 75.8%, a considerably stronger decline than the 0.5 pp. decline (to 62.5%) recorded for women. In order to reach the Europe 2020 goal of 75%, the male and especially the female employment rate will have to grow considerably until 2020. Graph 15 shows the development of the employment rate
for the 15 to 64 years old, the indicator used in
the Lisbon strategy(7), over the period 2000 to
2008. The picture resembles that for the 20 to 64 years old but on a about 5 pps. lower level. With a total employment rate of 64.6% in 2009, the Lisbon target of 70% has been missed by a considerable margin.
(7) The indicator has been refined by narrowing the age
bracket from 15-64 years to 20-64 years. For the 15-19 ear old the goals of a higher employment rate and higher education participation were inconsistent. Therefore, the -18% reduction in the number
of temporary employees (-0.9 Mio) accounts for 90% of the reduction in the number of all employees (-1.0 Mio; -6%).
The share of temporary employees differs strongly by age. In 2009, 40.2% of young employees had a temporary contract
(+0.2 pp. compared to 2008), against 11.5% (-0.5 pp.) among the 25-49 years old and 6.5% (-0.1 pp.) among the 50-64 years old. It is somewhat surprising that the share of temporary employees increased among the young during the crisis. The likely reason is that among the newly hired young the share of temporary contracts increased even further.
Table 8 - Structure of Employment by gender and contract type
2009
(in 1000s) 2008-2009Change 2007-2008Change Average change 2000-2009
Employees 180,150 -1.8% 1.6% 1.1% of which men 94,442 -2.9% 1.1% 0.6% of which women 85,708 -0.6% 2.0% 1.6% Self-employed 21,188 -0.4% -0.8% 1.3% of which men 14,161 -0.7% -1.4% 1.0% of which women 7,027 0.3% 0.3% 1.8% Full-time 175,043 -2.4% 1.4% 0.8% of which men 108,024 -3.1% 1.0% 0.5% of which women 67,019 -1.2% 2.3% 1.2% Part-time 38,769 1.1% 1.7% 2.6% of which men 8,683 3.0% 2.2% 3.2% of which women 30,085 0.5% 1.6% 2.4% Temporary Employees 24,158 -5.9% -1.4% 2.0% of which men 11,906 -7.3% -3.1% 1.4% of which women 12,252 -4.5% 0.2% 2.6%
Graph 14 - Employment Rate (age 20-64) 0.5 0.55 0.6 0.65 0.7 0.75 0.8 Europe 2020 target Employment rate (women age 20-64) Employment rate (men age 20-64) Employment rate (age 20-64) 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000
Source: Commission services.
Graph 15 - Employment Rate (age 15-64)
0.5 0.55 0.6 0.65 0.7 0.75 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Lisbon target Employment rate (women age 20-64) Employment rate (men age 20-64) Employment rate (age 20-64) Source: Commission services.
Table 9 provides an overview of the developments of the employment rates by gender, age and education levels. The two trends by gender and age, already discussed in section 1.6.1, are confirmed here. First, women do better during the crisis than men. Second, the development is particularly bad for the
young (men -3.3 pps., women -1.6 pps.), quite bad for prime age men (-2.3 pps.) and only slightly bad for prime age women (-0.6 pp.). The employment rate of old men remains nearly unchanged (-0.2 pp.), whereas the employment rate of old women still manages to grow by 1.0 pp., even during the crisis.
Table 9 - Employment Rates
Gender Employment Rate 2009 2008-2009 pps.Change 2007-2008 pps.Change 2000-2009 pps.Avg. Change
Total (age 15-64) male 70.7% -2.1 0.3 0.0
female 58.6% -0.5 0.8 0.6
Young (age 15-24) male 37.2% -3.2 0.0 -0.3
female 33.1% -1.5 0.3 -0.1
Prime age (age 25-54) male 84.6% -2.3 0.1 -0.1
female 71.7% -0.6 0.9 0.6
Old (age 55-64) male 54.8% -0.2 1.1 0.9
female 37.8% 1.0 0.9 1.2
Low education male 54.9% -3.0 -0.5 -0.6
female 37.7% -1.0 -0.5 -0.1
Medium education male 74.8% -2.2 0.5 0.0
female 63.1% -0.9 0.4 0.2
High education male 86.3% -1.2 0.2 0.0
female 79.8% -0.7 0.0 0.2
By gender and education we can see falling employment rates for both genders in all education groups. For men, the decline in the employment rate is much smaller for the highly educated than for the low and medium educated. For women, the employment rate declines for all three education groups by around 1 pp. It is noteworthy that the employment rates of highly educated men and women decline by around 1 pp., despite the increase in their absolute employment levels (see above, section 1.6.1). This shows that the increase in the number of highly educated has been even faster than the increase in the employment of highly educated.
1.6.3. Unemployment and participation The economic crisis had severe consequences for the number of unemployed (see Table 10). In 2009, the number of male unemployed grew by 36%, while the number of female unemployed grew by 19%. This again shows that women fared better in the labour market during the crisis. Looking at the age dimension, unemployment grew in all age brackets, with not very pronounced differences. For young and old men, the number of unemployed grew by around 30% and for prime age men by around 40%. For women, the number of unemployed grew in all three age brackets by around 20%. Table 10 - Unemployment
Gender (in 1000s or %)2009 2008-2009 Change (in % or pps.) Change 2007-2008 (in % or pps.) Avg. Change 2000-2009 (in % or pps.)
Total unemployment (age 15-64) male 11,658 36.2% 0.9% 1.5%
female 9,543 19.4% -3.4% -0.9%
Young unemployment (age 15-24) male 2,955 29.8% 2.6% 1.3%
female 2,196 17.3% -2.3% -1.1%
Prime age unemployment (age 25-54) male 7,595 39.7% 0.9% 1.6%
female 6,613 19.9% -3.6% -1.1%
Old unemployment (age 55-64) male 947 31.1% -3.4% 1.8%
female 579 22.2% -4.5% 2.7%
Long-term unemployment male 31.8% -5.0 -6.2 -1.5
female 34.8% -2.4 -5.3 -1.4
Source: Commission services.
The share of long-term unemployed fell in 2009, by 5 pps. (to 32%) for men and by 2 pps. (to 35%) for women. Per se, a falling share of long- term unemployment is good news. However, during downturns lower shares of long term unemployed are a common statistical effect due to the fact that many new unemployed enter the pool of unemployed. Yet, there is a risk of an increasing long-term unemployment if the probability of exiting unemployment do not pick up from the low levels achieved in many countries during the recession quarters.
The development of the labour force (Table 11) is by far less affected by the crisis than either employment of unemployment. Since the labour force is the sum of opposite employment and unemployment movements within the labour force itself, these two affect only marginally the latter, which is thus mainly the result of movements in and out of the labour force itself.
This is confirmed by Table 11, showing that some long-run trends basically continued in 2009. First, the labour force gets more female. The female labour force grew by 0.8% in 2009, while the male labour force shrank slightly by 0.1%. Second, the labour force gets older. The young labour force shrank by 3% (men) and 2% (women). The prime age labour force remained almost unchanged for men (-0.1%) and grew slightly for women (+0.6%). The older labour force recorded a strong growth, +2.8% for men and +5.2% for women. Third, the labour force gets more educated. The low-skilled labour force shrank considerably by 2.4% for men and 1.9% for women. The medium-skilled labour force shrank slightly, by 0.4% for men and 0.3% for women. On the other hand, the high-skilled labour force grew strongly. In 2009, there are 3.2% more high-skilled men and 4.9% more high skilled women as compared to the year before.
1.6.4. Unemployment rate and participation rate
The rise in unemployment documented in the last subsection can also be seen in the development of the unemployment rates (Table 12). The unemployment rate has grown stronger for men (+2.4 pps.) than for women (+1.3 pps.) and is now higher for men (9.1%) than for women (8.9%). Regarding age, the
young have larger unemployment rate increases than the older age brackets; regarding education, the low-skilled have higher increases than the high skilled. So, the largest increases in the unemployment rate are observed for young men (+5.3 pps.) and for low skilled men (+3.8 pps.). The smallest increases are observed for old women (+0.8 pp.) and for high skilled women (+0.9 pp.).
Table 11 - Labour Force
Gender (in 1000s)2009 2009-2008Change 2007-2008Change Avg. Change 2000-2009
Total (age 15-64) male 128,405 -0.1% 0.7% 0.5%
female 106,679 0.8% 1.3% 1.1%
Young (age 15-24) male 14,169 -3.0% -0.2% -0.6%
female 11,854 -2.0% -0.1% -0.8%
Prime age (age 25-54) male 97,274 -0.1% 0.4% 0.3%
female 82,413 0.6% 1.1% 1.0%
Old (age 55-64) male 16,963 2.8% 3.6% 3.2%
female 12,412 5.2% 3.4% 5.1%
Low education male 32,874 -2.4% -1.2% -1.3%
female 23,317 -1.9% -2.6% -1.8%
Medium education male 63,738 -0.4% 0.7% 1.0%
female 51,248 -0.3% 0.9% 1.2%
High education male 31,430 3.2% 2.8% 2.9%
female 31,863 4.9% 5.1% 4.8%
Source: Commission services.
Table 12 - Unemployment Rate
Gender Unmeployment rate 2009 in % 2008-2009 in pps.Change 2007-2008 in pps.Change 2000-2009 in pps.Avg. change
Total (age 15-64) male 9.1% 2.4 0.0 0.1
female 8.9% 1.3 -0.3 -0.2
Young (age 15-24) male 20.9% 5.3 0.4 0.4
female 18.5% 3.0 -0.3 -0.1
Prime age (age 25-54) male 7.8% 2.2 0.0 0.1
female 8.0% 1.3 -0.4 -0.2
Old (age 55-64) male 6.5% 1.4 -0.4 -0.1
female 5.9% 0.8 -0.4 -0.2
Low skilled male 14.8% 3.8 1.0 0.4
female 14.9% 2.5 0.2 0.1
Medium skilled male 8.3% 2.3 -0.3 0.0
female 8.5% 1.3 -0.6 -0.3
High skilled male 4.8% 1.4 -0.1 0.1
female 5.2% 0.9 -0.2 -0.1
The development of the labour force participation rates (Table 13) shows no uniform trend during the crisis. It fell for some socio- demographic groups but grew for others. For men it fell by 0.2 pp. to 77.8% and for women it grew by 0.4 pp. to 64.3%. For the young we see a declining participation rate especially for men (-0.9 pp.) but also for women (-0.5 pp.). In the prime age range the participation declined slightly for men (-0.2 pp.) and grew for women (+0.5 pp.). For the old we see a considerable increase for men (+0.7 pp.) and even more so for women (+1.4 pps.) which is in line with the long-run trend. By skill level and gender, we see a falling participation rate for low- (-0.5 pp.) and medium-skilled (-0.3 pp.) men, where as
the participation rate of high skilled men is almost constant. For women, the participation rates conditional on education are basically unchanged for all three skill levels, low, medium and high. It is noteworthy that this has been true for the whole decade. The female participation rates conditional on education barely changed during the 2000s. This means that the long-run increase in the unconditional female participation rate (on average 0.5 pp. per year during the decade) is almost completely due to a changing education composition of the female population. The share of low-skilled with low participation rates decreases and the share of high-skilled with high participation rates increases.
Table 13 - Labour Force Participation Rate
Gender Participation rate 2009 in % Change 2008-2009 in pps. Change 2007-2008 in pps. 2000-2009 in pps.Avg. change
Total (age 15-64) male 77.8% -0.2 0.3 0.1
female 64.3% 0.4 0.6 0.5
Young (age 15-24) male 47.0% -0.9 0.3 -0.2
female 40.6% -0.5 0.3 -0.1
Prime age (age 25-54) male 91.8% -0.2 0.1 0.0
female 78.0% 0.5 0.6 0.5
Old (age 55-64) male 58.6% 0.7 0.9 0.9
female 40.2% 1.4 0.7 1.2
Low skilled male 64.5% -0.5 0.1 -0.3
female 44.3% 0.1 -0.5 -0.1
Medium skilled male 81.5% -0.3 0.1 -0.1
female 69.0% 0.0 0.0 0.0
High skilled male 90.6% 0.1 0.0 0.1
female 84.2% 0.1 -0.2 0.1
Source: Commission services.
Next we want to investigate whether the labour market has been relieved during the crisis by lower labour force participation. Since the labour force participation rate approximately equals the employment rate plus
the unemployment rate (pr≈er+ur), a decrease
in the employment rate (e.g. during a recession) does not necessarily lead to an increase in the unemployment rate of the same size, insofar as the participation rate decreases as well. In order to analyse this question we look at the change in the employment rate and the change in the participation rate for different demographic groups since the first quarter of 2008. Graph 16 and Graph 17 show the difference in the employment rate and participation rate between the first quarter of 2008 and following quarters.
This approach is inspired by Elsby, Hobijn and Sahin (2010). Taking as an example the top left picture in Graph 16, men aged 15 to 24, we can see that between 2008Q1 and 2009Q4 the employment rate of males aged 15-24 fell by 4.4 pps., while their participation rate first stayed constant for three quarters and then fell by 1.3 pps.. Hence, in this group the fall in the participation rate by 1.3 pps. resulted in an increase of the unemployment rate by only 3.1 pps., despite a fall in the employment rate of 4.4 pps..
Graph 16 and Graph 17 show the development of employment and participation rates for the three main age and education groups divided by gender. They show that in three groups,
young men, young women and low-skilled men, a fall in the participation rate by around 1 pp. prevented a steeper increase in the unemployment rate. In the remaining groups, the fall in the participation rate was smaller, or it even increased. Where the participation rate declined, it did so with a delay of three to five quarters with respect to the decrease in the employment rate. This delay is observed in the US as well. The decrease in the participation rate for the young is likely to be explained by a postponed entry into the labour market (in the best case by staying in the education
system) rather than actually leaving the labour force. In several cases, the participation rate remains basically unchanged (high- skilled men, low-, medium- and high-skilled women), which means that the decreases in employment rates are not dampened by falling participation rates and translate one-to-one into increases in unemployment rates. Finally, another interesting group is that one of older men. Their employment rate stays basically unchanged during the period. Their increasing unemployment rate is entirely due to an increasing participation rate.
Graph 16 – Employment Rates and Participation Rates by Age
Participation Rate Employment Rate 2009q4 2009q3 2009q2 2009q1 2008q4 2008q3 2008q2 2008q1 -5 -4 -3 -2 -1 0 1
Change in Percentage Points
Men age 15-24 Participation Rate Employment Rate 2009q4 2009q3 2009q2 2009q1 2008q4 2008q3 2008q2 2008q1 -5 -4 -3 -2 -1 0 1
Change in Percentage Points
Women age 25-54 Participation Rate Employment Rate 2009q4 2009q3 2009q2 2009q1 2008q4 2008q3 2008q2 2008q1 -5 -4 -3 -2 -1 0 1
Change in Percentage Points
Men age 25-54 Participation Rate Employment Rate 2009q4 2009q3 2009q2 2009q1 2008q4 2008q3 2008q2 2008q1 -5